Solving sparse linear systems is a problem that arises in many scientific applications, and sparse direct solvers are a time consuming and key kernel for those applications and for more advanced solvers such as hybrid direct-iterative solvers. For this reason, optimizing their performance on modern architectures is critical. The preprocessing steps of sparse direct solvers, ordering and block-symbolic factorization, are two major steps that lead to a reduced amount of computation and memory and to a better task granularity to reach a good level of performance when using BLAS kernels. With the advent of GPUs, the granularity of the block computation became more important than ever. In this paper, we present a reordering strategy that increas...
In order to express parallelism, parallel sparse direct solvers take advantage of the elimination tr...
La résolution de systèmes d'équations linéaires creux est au cœur de nombreux domaines d'application...
As the computational power of high performance computing (HPC) systems continues to increase by usin...
Solving sparse linear systems appears in many scientific applications, and sparse direct linear solv...
Solving sparse linear systems is a problem that arises in many scientific applications, and sparse d...
We consider the solution of large sparse linear systems by means of direct factorization based on a ...
International audienceSolving sparse linear systems is a problem that arises in many scientific appl...
The cost of the solution phase in sparse direct methods is sometimes critical. Itcan be larger than ...
Sparse direct solvers using Block Low-Rank compression have been proven efficient to solve problems ...
High performance sparse direct solvers are often a method of choice in various simulation problems. ...
In the context of this thesis, our focus is on numerical linear algebra, more precisely on solution ...
International audienceKrylov methods such as GMRES are efficient iterative methods to solve large sp...
The memory usage of sparse direct solvers can be the bottleneck to solve large-scale problems involv...
This work, performed within the PETALh project, relates our attempts to improve the numerical behavi...
Controlled-source electromagnetic (CSEM) surveying becomes a widespreadmethod for oil and gaz explor...
In order to express parallelism, parallel sparse direct solvers take advantage of the elimination tr...
La résolution de systèmes d'équations linéaires creux est au cœur de nombreux domaines d'application...
As the computational power of high performance computing (HPC) systems continues to increase by usin...
Solving sparse linear systems appears in many scientific applications, and sparse direct linear solv...
Solving sparse linear systems is a problem that arises in many scientific applications, and sparse d...
We consider the solution of large sparse linear systems by means of direct factorization based on a ...
International audienceSolving sparse linear systems is a problem that arises in many scientific appl...
The cost of the solution phase in sparse direct methods is sometimes critical. Itcan be larger than ...
Sparse direct solvers using Block Low-Rank compression have been proven efficient to solve problems ...
High performance sparse direct solvers are often a method of choice in various simulation problems. ...
In the context of this thesis, our focus is on numerical linear algebra, more precisely on solution ...
International audienceKrylov methods such as GMRES are efficient iterative methods to solve large sp...
The memory usage of sparse direct solvers can be the bottleneck to solve large-scale problems involv...
This work, performed within the PETALh project, relates our attempts to improve the numerical behavi...
Controlled-source electromagnetic (CSEM) surveying becomes a widespreadmethod for oil and gaz explor...
In order to express parallelism, parallel sparse direct solvers take advantage of the elimination tr...
La résolution de systèmes d'équations linéaires creux est au cœur de nombreux domaines d'application...
As the computational power of high performance computing (HPC) systems continues to increase by usin...